See and Avoid Using Onboard Computer Vision

Abstract

An unmanned aerial vehicle (UAV), also known as a drone, refers to a pilotless aircraft, a flying machine without an onboard human pilot or passengers. As such, 'unmanned' implies the total absence of a human who directs and actively pilots the aircraft. Control functions for unmanned aircraft may be either onboard or off-board (remote control). That is why the terms remotely operated aircraft (ROA) and remotely piloted vehicle (RPV) are in common use as well [1]. The term UAV has been used for several years to describe unmanned aerial systems. Various definitions have been proposed for this term, like [2]:A reusable 1 aircraft designed to operate without an onboard pilot. It does not carry passengers and can be either remotely piloted or pre-programmed to fly autonomously.Recently, the most reputable international organizations -like the International Civil Aviation Organization (ICAO), EUROCONTROL, the European Aviation Safety Agency (EASA), the Federal Aviation Administration (FAA) -as well as the US Department of Defense (DoD), adopted unmanned aircraft system (UAS) as the correct official term. The changes in acronym are caused by the following aspects:1 The characterization reusable is used to differentiate unmanned aircraft from guided weapons and other munitions delivery systems.

Detection of aircraft below the horizon for vision‐based detect and avoid in unmanned aircraft systems

“…Importantly, our implementation achieves an overall frame rate that enables realtime performance on this desktop computer for window lengths less than 7 (15 Hz is desirable for real-time performance). Similar to previous observations made by Lai et al 28 and Mejias et al, 27 the performance of our CUDA implementations will scale with future improvements in GPU hardware.…”

Characterization of Sky‐region Morphological‐temporal Airborne Collision Detection

“…()]. In our inertial‐based ego‐motion compensation technique, aircraft yaw, pitch, and roll angles recorded by the onboard GPS‐INS system were used to offset apparent motion in images due to rotation of the camera sensor (Lai et al., ).…”

A Laser Obstacle Warning and Avoidance system for Manned and Unmanned Aircraft

“…Low visibility due to bad weather or natural/man-made obscurants is the main responsible for the steep increase of collisions with fixed ground obstacles in low-level operations of both UA and manned aircraft. While the development of an effective and certifiable Sense-and-Avoid (SAA) capability is considered paramount, significant research efforts are being devoted to the development of cooperative, non-cooperative and hybrid architectures capable of achieving this fundamental goal [1][2][3][4][5][6][7]. Research and Development (R&D) activities on laser-based obstacle detection and avoidance systems have progressed significantly over the past twenty years [8][9][10][11][12][13].…”